Project Summary Many students enter college with an interest in studying science. However, after enrolling in several college- level science courses, many pursue other majors. This phenomenon, referred to as a `leaky pipeline,' contributes to concerns over shortages in the biomedical sciences workforce, particularly among women, underrepresented minorities (URMs), and first-generation (FG) college students. Research indicates that early undergraduate research experiences and enriched curricular opportunities can encourage the pursuit of biomedical research careers. However, we still know very little about the underlying reasons why such programs are successful. To inform interventions aimed at repairing this leaky pipeline, it is critical to identify changes in psychological processes associated with both attrition and retention, and to examine how we might apply knowledge of these processes to support the pursuit of biomedical careers. This renewal application utilizes and supplements data collected during the original funding period to investigate how various types of undergraduate enrichment experiences support the pursuit of biomedical research careers and to identify the psychological processes that underlie such support. Specific Aims: (1) Examine how early undergraduate enrichment experiences support initial (end of sophomore year) engagement in the biomedical sciences through changes in underlying psychological processes (science self-efficacy, interest/value, identity, and achievement goals), (2) Examine how the type, timing, and dosage of undergraduate enrichment experiences support sustained (two years post-college) engagement in the biomedical sciences through changes in underlying psychological processes. For both aims, differences in the effects of these experiences among women, URMs, and FGs will be investigated. Four cohorts of Duke University students from the original grant will continue to be surveyed longitudinally from their first year in college to two years after graduation (the first cohort graduated from college 2 years ago and the final cohort are currently college juniors). In the current application, these data will be supplemented by (a) adding ~350 comparison group Duke students to the original longitudinal sample from Duke, (b) adding one new cohort of Michigan State University students who will be followed from their 1st year through one year post graduation, and (c) adding a qualitative component using in-depth interviews with ~80 participants. The result will be a large, relatively diverse dataset (N = 2,293; ~59% female; ~21% URM; ~19% FG) with detailed psychological and behavioral measures taken each year for 6 years supplemented with qualitative interviews. IMPACT: This rich longitudinal, multi-method dataset will provide key insights into the types of undergraduate enrichment experiences that encourage persistence in biomedical sciences, especially among women, URMs, and FGs, thus serving as a model for more effective allocation of resources throughout U.S. universities.